OncoML: A Multi-omics-based Ensemble Learning Approach for Targeted Cancer Drug Prediction

Overview

Predicting the response of a cancer patient to a specific drug is a major goal in precision oncology. The current approach to cancer treatment is a one-size-fits-all approach, failing to comprehend tumor heterogeneity, results in 75% ineffectiveness of cancer treatment. The availability of large computing cycles has allowed the creation of high-throughput genomic technologies and large-scale sequencing studies including The Cancer Genome Atlas (TCGA). Recent research has focused on modeling of drug prediction by applying machine learning on genetic mutations or using microRNA (miRNA), a key biomarker of cancer, on mice models or cancer cell lines (Kalamara et al., 2018). Although these approaches demonstrate improved potential of targeted drug prediction, they present some limitations. Gene mutations have shown to account for a subset of candidate biomarkers, while miRNA-based gene expression is regarded as offering more predictive modalities; both can be complemented by the multi-omic view of cancer. The integration and analysis of these multi-omics data is a critical step to deliver on the promise of precision oncology: selecting drugs for patients based on their individual data. The solution is a machine learning platform that analyzes genetic mutations, miRNA, and other pharmacogenomic data of various cancer types of real patients and predicts targeted drugs for any given cancer patient with a high accuracy. This approach of using machine learning on multi-omics cancer data of real patients from The Cancer Genome Atlas to deliver targeted cancer drug therapy is superior to existing mono therapy research on cancer cell lines. PENNSYLVANIA

Competition history

• JSHS 2022 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.